Artificial Neural Network Based Rotor Flux Estimation And Fuzzy-Logic Sensorless Speed Control Of An Induction Motor

Abstract

Induction motor is one of the most widely used machines in industrial applications due to its high reliability, relatively low cost, and less maintenance requirements. Variable speed induction motor drives that operate without speed or position sensors have the benefits of reduced drive system’s size and overall cost as well as high system reliability. This thesis aims to design rotor flux estimation based on Artificial Neural Network and fuzzy-logic based speed control of indirect field-oriented control for an induction motor drive to achieve high dynamic performance. For implementation purposes, the corresponding model is done using MATLAB software tools. The performance of speed control of the indirect vector control of an induction motor for conventional PI and Fuzzy logic speed control has been compared. Accordingly, the PI speed controller has a settling time 0.2sec, 5.851% overshoot, 0.045sec rise time, and has no steady-state error for 100 rad/s at no-load condition. On the other hand, the fuzzy logic speed controller has a settling time of 0.085sec, 0.505% overshoot, 0.0386sec rise time, and has no steady-state error for 100 rad/s at noload condition. Thus, values show that the fuzzy-logic based speed control of indirect vector control of induction motor gives a better speed response than that of the conventional PI controller. The performance of PI and fuzzy logic speed controller evaluated under the application of sudden load change for constant setpoint conditions. The settling time, rise time, and peak time values are increased when compared to the system simulated under noload condition. This shows that when the load is increased, the system performance will be decreased. There is no variation in the case of overshoot but there is a steady-state error in the result with the system simulated under no-load condition. Generally, the rotor flux angle is estimated by using ANN and the speed estimated by using open-loop speed estimation techniques. The sensorless speed control tracks the actual speed of the motor when the motor run at the constant speed, changing the reference speed, and in the regenerative mode of operation.